Note: Descriptions are shown in the official language in which they were submitted.
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 3/39
Background of the invention
Field of the invention:
The present invention relates to high quality recording, and more specifically
to
a high quality Membrane-less Microphone capable of functioning in a very wide
range of frequencies without distortions, which can be at the same time very
compact and much cheaper than the state-of the-art high-end microphones. In
some
of the embodiments it can also be easily made directional or even very
directional,
and it is much less affected by electromagnetic interference. In other
embodiments
similar principles can be used for reproducing sounds.
Back. or~und
Unlike computers, the field of Iii-Fi recording has advanced much more slowly,
so
that for example some systems existed even 20 or more years ago with quality
not
very different or even better than many systems that are sold today. One of
the
elements that hasn't changed much for example is the Microphone. .Although
narmal microphones are very cheap, their range is usually limited up to around
lOKHz and is typically not free from various distortions. Other microphones
that
can reach 20 KHz or close to it typically cost tens or hundreds of dollars and
still
have various limitations, and higher-end microphones, for example of the types
needed for Live Music performance or for the Mass media broadcasting, such as
for
example Radio or TV, are typically much more expensive and can cost even
thousands of dollars. The main .reason for these limitations is the fact that
normal
Microphones use a membrane, which is a mechanical element, and therefore they
are limited by the mechanical qualities of the membrane. In addition, normal
dynamic electronic microphones use a sensitive coil that is affected by the
movement of a magnetic element that is attached to the membrane, so the weak
currents in the coil need significant amplification, and this coil is
therefore
susceptible to electromagnetic interference, for example in cars, and even
shielding
it with a metal mesh only partially solves the problem, because for example it
cannot be fully electromagnetically shielded in the direction where the sound
needs
to come in. On the other hand, condenser microphones, which are based on a
changing capacitor instead of a coil, typically have a more flat response on a
wider
frequency range than dynamic microphones, but they are still limited by the
physical
properties of the membrane, they are typically much more expensive, and they
typically suffer from much less tolerance to loud noise saturation. Even
Passive
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 4/39
reflective Optical Microphones, that have the advantage of being immune to
electromagnetic interference, still need to use a membrane and thus still have
the
mechanical limitations imposed by the membrane itself, such as bandwidth
limitations and varying response curves that depend on the frequency. Speakers
(for
example in earphones, but especially loudspeakers, suffer from very similar
limitations, because they are typically dependent for sound reproduction on
one or a
few relatively large membranes.
Summary of the invents~n
The present invention discloses a novel type of microphone that is able to
detect
directly sound waves in the air without a membrane, and therefore has the
advantage
that it can function without distortions or at least with much less
distortions than
conventional microphones over a much wider range of frequencies, such as for
example between 0 to 500KHz or other desirable ranges, giving a preferably
more
or less flat response, so that the sensitivity function is preferably similar
over the
entire range of frequencies. In addition, it does not use an electromagnetic
sensing
coil, so it is much less susceptible to electromagnetic interference. In
addition, at
least in some of the embodiments shown. it can be made directional or even
very
directional as needed, and thus can be used for example in noisy environments.
It
can also be much more robust than ordinary microphones in being able to handle
even very low sound levels up to very high sound levels, and can also have a
very
fast transient response. This is preferably accomplished in the following
preferable
ways:
1. The sensing of sound is preferably based on sensing the interferences or
distortions that audible or higher sound waves create on one or more base
signals
that are preferably of a considerably higher frequency than audible sounds,
such
as for example a few hundred KIIz or even 1 or more ll~IHz. Unlike usual
ultrasound sensing, where the signal is emitted and reflected back to the same
place, preferably the sensor and the detector are separate. t~nother possible
variation is that the same device is used both for generating the signal and
for
detecting back the altered signal. I~owever, since unlike usual ultrasonic
sensing,
the sensed target is sound waves in the air itself, there is no normal
reflection
from the sensed target back into the emitting element. Therefore, preferably
sensor is on the other side after the signal has gone through the air, or for
example the signal is reflected back to the direction of the emitter of the
ultrasonic signal by a constant reflector, in which case any distortions
caused by
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 5/39
the reflector itself are preferably taken into account and ignored by the
decoding
algorithm, so this variation is less desirable, and also in this variation the
signal
goes twice through the same air gap, which has to be taken into consideration.
Preferably the air gap between the transmitter and the receiver is small
enough to
detect just 1 peak of the wave, so that for example if the desired detectable
frequency range is for example up to 20KHz, preferably the gap is 1.7 cm or
less, and if the desired detectable frequency is up to 70KHz, preferably the
gap is
5mm or less. On the other hand, since the smaller gap contains also less peaks
of
the ultrasound wave, preferably the ultrasound frequency used is as high as
possible in order to improve the resolution or sensitivity. The Ultrasonic
signal
or signals can be generated and/or detected for example by a Quartz crystal,
or
by a Piezoelectric ultrasonic sensor, or for example by newly available MEMS
(Micro-Electro-Mechanical-Systems) silicon-based ultrasonic sensors (such as
for example by Sensant Corp.), or by any other known means for creating and/or
detecting ultrasonic waves. The new MEMS sensors have the advantage that
they are more efficient at transferring electrical energy into acoustic
energy, and
they can be a 10,000 times more sensitive than comparable piezoelectric
sensors.
Also, they can work for example in the range of 200I~Hz-SMHz, compared to
Piezoelectric devices, which typically work only in the range of 50-200KHZ,
and also they can be cheaper and smaller than piezoelectric sensors, so
preferably such MEMS are used at the highest possible frequency. (Actually the
MEMS for example do use very little micron-scale membranes for the sensing
and the transmitting, so at least in the embodiments that use them the
microphone of the present invention is not entirely without a membrane,
however they are used very differently that an ordinary membrane in a
microphone to detect sounds indirectly). In case a Quartz crystal is used, its
own
natural frequency is preferably used as a base reference. The transmitted
ultrasound signal can be of any desired shape, such as for example a sine
wave,
and it can be for example a consecutive signal or for example based on very
short pulses. The ultrasound beam or beams are preferably emitted all the time
that the microphone is turned on. Another possible variation is that they are
activated for example only when the microphone senses that any sound is
available, for example by using at least one transmitter-sensor pair that is
always
active when the Microphone is on, or by using for example some other
preferably small membrane for sensing that there is any sound activity.
Another
possible variation is that one or more pairs are always active when the
microphone is on but on a very low level, and the level of the ultrasonic beam
is
immediately increased when the microphone senses that any relevant sound is
CA 02446344 2003-10-23
23110103 Boris Dechovich & Yaron Mayer 6/39
entering the system. Preferably the decoding of the detected interference is
based
on phase shifting detection, and this can be converted back to the detected
sound
frequency for example by deleting the phase-shifted signal from the base
reference frequency, and/or using an interferometer, and/or by using a
feedback
loop that changes the transmitted frequency, and/or by any other known means.
Another possible variation is to detect for example in addition or instead
also
frequency shifting and/or amplitude distortions and/or any other distortions.
The
decoded signals can be for example analogue or digital, but the digital
embodiment is more preferable because processing can be more easily done with
a digital processor and because a digital signal that is transmitted from the
microphone on an electrical wire is mare immune to electromagnetic
interference, even though the wire is preferably shielded in any case. If an
analogue signal is used then it is preferably transmitted for example with
frequency modulation or PWM (Pulse Width Modulation), in order to make it
more immune to electromagnetic interference. One possible variation is using
for
example at least two baseline high-frequency signals, one that is isolated
from
the sound and one that is exposed to the sound, so that they can be compared.
Another possible variation is using for example only one frequency that is
exposed to the sound and comparing it for example to a digital representation
of
the original undistorted base high frequency that is for example pre-stored is
the
processor's memory. If for example 1 MHz is used as the base frequency, this
can be used for example for detecting sound signals of up to SOOKHz, which is
half the wavelength, however, as explained above, preferably the ultrasound
frequency is as high as possible. Another possible variation is to use for
example
as a reference baseline the normal interference pattern between one or more
ultrasonic signals, and detect distortions as deviations from this
interference
pattern caused by normal sound waves in the air. Another possible variation is
to
use for example optical signals instead of the ultrasound signals, and detect
for
example the distortions that the varying air pressures (caused by the sound
waves) have for example on an interference pattern of two or more light beams,
so that the deviations of the normal interference pattern are detected, or for
example detect the small Dopier shifts that this can cause. This variation
might
be called for example an optical microphone without a membrane. Another
possible variation is to trap for example some preferably very small particles
or
for example ionized gas inside some enclosure and thus measure the changes in
light caused by the movements of these minute particles. Another possible
variation is to similarly use for example other types of frequency, such as
for
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 7/39
example very high electromagnetic frequency. Of course various combinations
of the above and other variations can also be used
2. Preferably the microphone is naturally at least partially directional, for
example
by putting the sensors inside a small acoustic tube, so that the tube itself
allows
more sounds to come in from its front than from its sides. Preferably the
Microphone can be made even more directional by using a number of sensors
and/or a number of high frequency sources inside the microphone, so that by
taking into account the differential effect on them, the direction of the
sound can
be determined, and sounds from unwanted directions can be cancelled out for
example by appropriate phase shifting. Another possible variation is to use
for
example this shifting in order to allow the user to electronically change the
level
of directionality and/or to electronically change the angle of input from
which
the sound is picked up. Preferably the default level of directionality is not
too
high, such as for example not less than a spread of 20 or 30 degrees, since
otherwise for example the user's movements can cause the speech to fluctuate
in
and out of focus. Another possible variation is to use for example a Fourier
transform in order to filter out the relevant directions. If MEMS sensors are
used,
the directionality control can be even easier, since each MEMS chip can use a
large array of such sensors, so data. from a number of different sensors can
be
used, for example even with the same transmitter-sensor pairs. Another
possible
variation is that the processor in this case is also integrated into the same
chip, or
for example the logic for the various processings needed by the microphone is
integrated into the chip, for example as an ASIC. Another possible variation
is
that for example at least one MEMS miniscule drum is used within each
transmitter and within each sensor, and they are arranged in pairs, and then
preferably the integrating logic is across these chips. Of course various
combinations of the above and other variations can also be used.
3. Preferably the microphone is very robust is terms of the range of volume
levels it
can detect, so that its high sensitivity allows it to detect audible sounds
from very
low levels up to high volume saturation. This is easily accomplished since in
the
embodiments that use for example the MEMS sensors these sensors can detect
even very slight distortions created in the ultrasonic signal, so they can
detect
also very low volumes, and on the other hand if even very high sound levels
are
used, this will create larger distortions of the ultrasound signal or signals
but will
not cause saturation problems, as can happen with conventional microphones
that use a membrane. In addition, due to the fast reaction of the ultrasound
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 8/39
sensors, the microphone can have also a very good transient response - which
means that it can react very quickly to sudden changes in the sound.
Preferably if
there is any non-linearity in response at certain frequencies and/or volumes
it is
automatically corrected electronically andlor digitally.
4. Another possible variation is to use a similar process for example in
reverse, so
that interference patterns created between two or more high frequency sources
can be used to create any lower-frequencies and volumes desired, thus creating
also a membrane-Less loudspeaker or earphone, preferably also with a broad
frequency range and less distortions than ordinary speakers. Preferably this
is
done with a large array or matrix of membranes, such as for example a MEMS
matrix, which preferably fills the surface area better than round membranes,
by
being for example rectangular or six-sided, like in a honeycomb, which is
indeed
the structure in the MEMS shown in Fig. la. Preferably the distortions used
are
based on phase shifting and/or frequency shifting and/or for example also
amplitude distortions, so that these shiftings create wave fronts which move
at
the frequency of the desired lower frequencies that need to be recreated (For
example, one speaker can broadcast at 60KHz and the other at 60KHz+ the
frequency range needed for the audible sound, for example 300-20,000 Hertz,
and the user hears only the difference between them as a result of the
frequency
mixing and interference). Preferably for this the minute membranes are
electronically connected in a row. Preferably the ultrasonic frequencies used
in
this case are close to the natural resonance frequency of the minute membranes
in order to increase the overall efficiency of the process. In this case
various
frequencies that are created are preferably mixed together for example within
a
hollow resonance box or for example a hyperbolic or parabolic reflector,
depending on whether the desired output is more directional or more omni-
directional. Another possible variation is that preferably either multiple
membranes are used with varying diameters, so that they can be vibrated mare
efficiently at a large range of frequencies, or for example large arrays of
minute
membranes, for example MEMS membranes, are used and vibrated at all desired
frequencies, with various combinations of vibrating membranes individually or
with preferably high synchrony among them. This way, the high frequencies can
be created for example very simply by vibrating the minute membranes, and
lower frequencies can be simulated for example by slowly vibrating a large
number of the minute membranes in synchrony, since each minute membrane
has only a very small displacement and for lower frequencies a larger
displacement is needed, thus creating a simulation ~f one large slowly
vibrating
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 9/39
membrane. This way for example any sets of large membranes, medium
membranes and/or small membranes or combinations thereof can be dynamically
simulated and changed in real time on the fly, for example with either
separate
sets of minute membranes for each size, or for example with at least partial
overlap of minute membranes across simulated sets, so that the membranes are
treated like a single large membrane but with much better flexibility and
freedom of movement of each part of it than a single large membrane. This has
the further advantage that very compact high level speakers (for example
loudspeakers or earphones) can be built this way, preferably with processor or
computer control. Preferably in all of the above variations the compact
speakers
are connected to an efficient external heat sink in order to efficiently get
rid of
the heat caused by the limited power conversion of any speakers. Another
possible' variation is to use similar principles for broadcasting sound to
various
desired places away from the speaker (for example for creating various
surround
effects without having to use mare speakers), by creating crossing points
between two or more directional ultrasound beams at the desired locations. By
changing for example the angle of the beams, the crossing point where the
sound
will be heard can be changed, and for example by changing the directionality
of
the ultrasound beams the size of the hearing area can be changed, and this
change can be done for example by the user and/or automatically by the system.
(Although there is for example a speaker by the British company lLtd, which
uses a flat panel with X54 small speakers, which produce tight focused beams
of
sound, which are distributed into the room and reflected off surfaces to
create a
multi-channel sound field, projecting normal sound can cause the sound to be
heard also out of the desired places. On the other hand, using ultrasound
interference can create much more precise effects in the desired spaced).
However, since the minute membranes can have only very small displacement,
another possible variation is to use instead of the membranes similar minute
elements that are preferably more rigid and are preferably not connected or
only
partially or loosely connected at their circumference to the frame that
surrounds
them and preferably reside above a longer cavity, so that they can have a
displacement range preferably even larger than their own size. So that for
example a lmm size rigid membrane can have a displacement range of for
example a few mm, as shown in Figs. 4a-b. This can be done for example by
connecting the more rigid free element to a small needle that goes through a
low
friction tunnel and at the other end of the needle is the part that is moved
for
example by an electromagnetic coil or more preferably by changes in a
capacitor
or for example by a Piezo element, or for example capturing the element freely
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron iVIayer 10/39
within a mesh or narrower passage that allows air to flow but keeps the
element
from escaping from its displacement path. Another possible variation is that
since the minute membranes are very directional and since in loudspeakers the
sound is preferably omni-directional, they are built for example on a wavy
and/or convex surface and/or are pointed at a hyperbolic reflector which
reflects
back the sound in much more directions. Another possible variation is that
more
than one layer of these structures is used, preferably in a configuration
allowing
free air flow, so that preferably the displacements are cumulative even in
each
small column. Another possible variation is that one or more larger structure
or
surface containing the small elements can be vibrated in order to create the
low
frequencies, and the higher ones are created by the minute elements or minute
membranes by any of the means described above. Another possible variation is
to use similar freely moveable, preferably rectangular or hexagonal, small
elements, preferably connected to one or more larger elements, which are
vibrated for example by one or more electromagnetic coils or capacitors or
Piezo
elements, so that the entire bunch of elements vibrate together without having
to
apply a separate electromagnetic coil ar capacitor or Piezo for each of them,
which can thus make the design even cheaper. This is somewhat similar to the
method of vibrating a single panel by the NXT (or NEXT) speaker technology
by the New T~arasducers company, however their technology is based on
creating standing waves and resonances in one relatively large panel, which
has
a wide frequency response, but still has a zigzaggy and not flat response
curve
because the single panel cannot really vibrate freely in all the range of
frequencies. In contrast, the above variation has the advantage that each
smaller
element can vibrate much more freely, without the above limitations, and thus
the combined vibrations of the entire structure create both lower and higher
frequencies at the same time with much less distortions. A few possible
variations of this are shown for example in Figs. Sa-g. Another possible
variation
is to use also a special electronic pre-correction, preferably in a drive
circuit or
DSP on the speaker itself, before vibrating the small elements, so that any
remaining distortions are electronically taken into account and fixed in
advance.
Another possible variation is for example to simply vibrate a large number of
small membranes or elements in synchrony for any frequencies. Another
possible variation is to use for example a normal type of speakers but to add
one
or more even smaller membrane than the usual tweeters, in order to display
sound well above 20KHz, however the problem is that most people have very
limited hearing above 20Ki-~z. Another possible variation is automatically
downshifting the higher frequencies, preferably digitally, so that for example
if a
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 11/39
guitar string can have harmonies that vibrate up to 70KHz, but humans can't
hear these frequencies anyway, these frequencies can be for example downward
converted, so that for example the range of 20-70KHz becomes correspondingly
converted upon replay to a range of for example 20-22KHz or less, so that it
preferably will appear to the listener as audible sounds at the high end of
the
pitch. Preferably the users have the ability to choose the desired range of
the
sound to be downshifted and/or preferably also the amount of downshifting
and/or the ratio of conversion (for example convert the 20-70KHz range down to
a range of 2KHz or a range of 3 KHz, etc). This has the huge advantage that
each
user can adjust the sound playback so as to optimize his ability to hear
according
to his own limitations, so that for example someone who can hear well only up
to 18KHZ will choose a lower downshifting and thus be able to hear fantastic
sounds he has never been able to hear before, for example with music playback.
Preferably this can be done either on the fly, for example in live concerts
(however, in this case the users need headsets if the adjustable embodiment is
used), and/or for example when playing back a recording. Another possible
variation is for example to similarly use automatic up-shifting of very low
frequency sounds, so that people can hear them more clearly at some higher
frequencies. With recording preferably the down-conversion is done either
during the recording or during the playback, or for example the user can have
a
choice about this, or some combination of the above. This can be used also for
example in a communication device for exchanging sound with Dolphins or
other animals that have a much different hearing and speaking range than ours.
So since Dolphins can for example easily emit and hear sounds between 20KHz
to 200KHz, another variation is to build for example a two-way automatic
conversion system so that sounds that dolphins emit for example between
20KHz to 200KHZ are down-converted for example to sounds between 0-
20KHZ, and sounds that the human emits back to the dolphins are up-shifted and
preferably spread for example to a range of 20KHz-200KHz. Like other features
of this invention, any of the variations for reproducing sounds described
above
can be used also independently of any other features of this invention.
Of course various combinations of the above and other variations can also be
used. This type of microphone can be very useful for example for recording
live
music, for high-quality recording for the mass media, such as for example TV
or
Radio interviews, and for noisy environments or environments where there are
electromagnetic interferences, such as for example in cars. This can be
especially
important for example for use with Telematics systems in cars, which are
interactive
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer ~2/~9
wireless information systems for cars which use interactive voice-based menus,
since computer systems that can recognize spoken words are very sensitive to
distortions, so normal microphones could be very problematic in cars both
because
of the noisy environment and because of their susceptibility to
electromagnetic
interference. An additional advantage is that apart from the high quality and
frequency range, it can be much more lightweight and compact than the state-of
the
art high-end microphones, so it can be much more convenient for example for a
reporter who does field-work and needs to report from outside. Other possible
applications are for example using this microphone with the more directional
variations in a personal cellular remote speaking unit, that can be positioned
for
example on a table at a distance of for example 1 meter or less from the user,
or for
example in a handheld cellular or mobile phone that can be held at a certain
distance
from the head (preferably in each of these cases together with a directional
speaker).
Another possible implementation is to use this microphone in better quality
non-
mobile phones or cellular phones or Internet phones. Although the frequency of
transmitted speech in both non-mobile and cellular phones is typically limited
to
about 3KHz because of frequency utilization considerations, newer phones or
cellular phones or Internet phones might be for example be based on TCP/IP and
High digital condensation ratio such as for example DSP-based MP3 or other
high-
condensation algorithms, and then higher frequencies might be used. Another
possible variation for example in normal phone lines or cellular phone lines
that are
using only a limited band such as for example a 3 KHz band, is for example to
automatically downshift the higher frequencies to a preferably small range at
the
high end of the limited band, in a way similar to the downshifting described
above
in clause 4. This way, high pitch sounds such as for example vowels with a lot
of 0-
crossings, and/or vowels such as for example "ch" or "s" or "n" or "b" or "p"
or "t"
will sound more clear on the phone, instead of the situation today that many
times
the users have to code these vowels into words when dictating for example an
exact
name. This can be used also independently of any other features of this
invention.
Another possible variation is to automatically up-shift back the downshifted
band at
the hearing side, and preferably also spread it back to the original wider
range at the
higher frequencies. Another implementation is using this microphone for
example
for various army uses, where sometimes frequencies of even up to a few hundred
KHz or more need to be detected, for example to monitor sounds created by
various
devices. In this case, preferably the detected sounds can be for example
displayed
visually, or downshifted to a more audible range, as explained above. Other
possible
applications are for example as an additional method for detecting seismic
activity
and/or sensing for example earthquakes, by sensing for example vibrations at
low
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 13/39
frequencies or infrasound. For playing back infrasounds, they can be displayed
for
example visually, or for example played back for hearing by automatically up-
shifting the frequency, and preferably also spreading it on a wider range. Of
course,
various combinations of the above and other variations can also be used.
Brief description of the drawinEs
Figs. la-b are illustrations of Sensant's miniscule ultrasound
emitters/sensors built
on the surface of Silicon.
Figs. 2a-b are illustrations of two possible variations of using transmitter-
sensor
pairs.
Fig. 3 is an illustration of a phase-shift view on a scope.
Figs. 4a-b are illustrations of two preferable examples of using, instead of
minute
membranes, a matrix of minute elements that are preferably more rigid and can
move more freely.
Figs. 5 a-g are illustrations of a few preferable variations of using freely
moveable,
preferably for example rectangular andlor hexagonal and/or fractal shaped,
small
elements, connected to one or more larger elements, which are preferably
vibrated
so that the entire bunch of elements vibrate together without having to apply
a
separate transducer for each of them (i.e. the elements can be for example of
the
same type or a mix of different types).
Important Clarification and Glossaryn
All these drawings are exemplary drawings. They should not be interpreted as
literal positioning, shapes, angles, or sizes of the various elements.
Throughout
the patent when variations or various solutions are mentioned, it is also
possible to use various combinations of these variations or of elements in
them,
and when combinations are used, it is also possible to use at least some
elements
in them separately or in other combinations. These variations are preferably
in
different embodiments. In other words: certain features off the invention,
which
are described in the context of separate embodiments, may also be provided in
combination in a single embodiment. Conversely, various features of the
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & ~'aron Mayer 14/39
invention, which are described in the context of a single embodiment, may also
be provided separately or in any suitable sub-combinations
Detailed description of the ureferred embodiments
All of the descriptions in this and other sections are intended to be
illustrative
examples and not limiting.
Referring to Figs. la-b, we show an illustration of Sensant's miniscule
ultrasound
emitters/sensors built on the surface of Silicon, as quoted from
http://www.sensorsma~.com/articles/0~00/17/main.shtn~I. As can be seen, the
Silicon sensors resemble tiny drums with a thin, ultrasensitive nitride
membrane that
vibrates to send and receive ultrasound. The membrane and the underlying
silicon
substrate form the top and bottom plates of a capacitor. Changes in the
voltage on
the capacitor displaces the nitride membrane, and displacements of the
membrane
cause detectable changes in capacitance. As explained above in the summary,
these
sensor and emitters have the advantage that they are more efficient at
transferring
electrical energy into acoustic energy, and they can be a 10,000 times more
sensitive
than comparable piezoelectric sensors. Also, they can work for example in the
range
of 200KHz-SMHz, compared to Piezoelectric devices, which typically work only
in
the range of 50-2001~HZ, and also they can be cheaper and smaller than
piezoelectric sensors. Preferably the air gap between the transmitter and the
receiver
is small enough to detect just 1 peak of the wave, so that for example if the
desired
detectable frequency range is for example up to 20KHz, preferably the gap is
1.7 cm
or less, and if the desired detectable frequency is up to 701~Hz, preferably
the gap is
Smm or less. On the other hand, since the smaller gap contains also less peaks
of the
ultrasound wave, preferably the ultrasound frequency used is as high as
possible in
order to improve the resolution or sensitivity. Since higher ultrasonic
frequency
means more ultrasonic peaks within the small needed gap between the
transmitter
and the sensor, preferably such MEMS are used at the highest possible
frequency. 1n
the embodiments that use such MEMS, one or more such miniscule drums can be
used for each transmitter and for each sensor, and r.he whole set of
transmitter-
sensor pairs can either reside for example on one special integrated MEMS
chip, or
for example each pair can reside on one MEMES chip, or for example each
individual transmitter and each individual sensor is based on one or more MEMS
elements. Another possible variation is to use for example more sensors than
transmitters or more transmitters than sensors. But preferably each sensor is
paired
with one transmitter and at least one such pair is used. Preferably the
ultrasonic
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 15/39
beam between them is very narrow and directional so as to increase the
efficiency
and avoid disturbances between pairs if more than one pair is used, as
explained
below in the reference to Figs. 2a-b. This is easy to accomplish since these
miniscule drums have a very directional beam and are very small. Of course
this is
just one example of a possible implementation, and many other possible
variations
of preferably minute ultrasonic emitters and sensors can also be used. Also,
as
explained in clause 4 in the patent summary, another possible variations is to
create
also wide-frequency band speakers by using for example an array or matrix of a
preferably large number of these drums and vibrate them at all desired
frequencies,
with various combinations of vibrating membranes individually or with
preferably
high synchrony among them. This way, the high frequencies can be created for
example by simply vibrating the minute membranes, and lower frequencies can be
simulated for example by slowly vibrating a large number of the minute
membranes
in synchrony, thus creating a simulation of one large slowly vibrating
membrane.
Preferably the number of membranes used changes gradually depending on the
frequency, so that the lower the frequency the more membranes are activated.
This
has the further advantage that very compact high level speakers can be built
this
way, preferably with processor or computer control. The minute membranes don't
have to be circle-shaped but can be also for example rectangular or with more
than 4
sides.
Referring to Figs. 2a-~, we show illustrations of two preferable variations of
using
transmitter-sensor pairs. Fig. 2b is a side view cross-section, in which the
microphone is in some depth inside an acoustic tube (20) and only one pair of
ultrasound transmitter (21a) and sensor (21b) is used. The acoustic tube
itself can
thus serve as a constrictive boundary, thus defining in general the shape of
sound
beam 24. Another possible variation is to use for example some parabolic sound
reflector around the pair or pairs instead of just a tube. Fig.. la is a top
view of a
preferable variation in which preferably 2 or more but more preferably at
least 3 (or
more) transmitter-sensor pairs are used (pair 21 with transmitter 21 a and
sensor 21 b,
pair 22 with transmitter 22a and sensor 22b, and pair 23 with transmitter 23a
and
sensor 23b). As shown in the illustration, preferably the pairs are arranged
so that
the directions of the beams do not interfere with the other pairs and
preferably the
distances among the pairs are bigger than the gaps within the pairs. The
sensors
and/or the transmitters can be for example suspended inside the microphone in
mid-
air for example by thin wires, so as not to obstruct the passage of lower
frequency
waves. Another possible variation is that in order to further reduce
disturbances each
ultrasound sensor and/or transmitter can be for example encased in some wider
envelope or for example some parabolic enclosure that absorbs or concentrates
any
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 16/39
residual parts of the ultrasound beam. Another possible variation is that each
pair is
within a hole in some surface so that there is more isolation between the
pairs,
however this has the disadvantage that waves with lower frequencies might be
able
to only partially penetrate these holes. By using a larger distance among the
3 pairs,
better directionality control can be obtained, as explained in above clause 2
of the
patent summary, and also the microphone can be even more optimal also for
lower
frequencies. Therefore, the microphone can be for example relatively flat, and
with
a diameter of for example a few centimeters or more or less. The pairs can be
for
example at the top surface of the microphone, so that no acoustic walls are
used to
create directionality, and then preferably all directionality is achieved by
the
electronics that takes into account the different reaction of the pairs
depending on
direction. This has the advantage that the microphone can be flexibly changed
from
almost omni-directional to very directional, however is has the disadvantage
that no
automatic directionality is added by the walls. Another possible variation is
that the
surface that contains the pairs is lower inside the acoustic walls of the
microphone,
or for example this surface is movable and is for example automatically
adjusted in
addition or instead of the electronically achieved directionality, when the
user
adjusts the directionality. Preferably at least 3 pairs are used in order to
achieve
proper directionality control, however of course more than 3 can also be used.
Another possible variation is to use for example some combination of Fig. 2a
and
Fig. 2b, so that for example both the top ~ pairs exist and one or more
surfaces of a
more inner pair or pairs also exist, and the microphone can for example
automatically choose which of the pairs or sets of pairs to use according to
the
directionality adjustments requested by the user. Another possible variation
is to use
for example a number of types of pairs within each surface or at different
surfaces,
or for example more sensors than transmitters, so that the farther sensors are
used
for sensing lower frequencies and the smaller pair gaps are used for sensing
higher
frequencies. This should be no problem since for example the MEMS sensors and
transmitters should be very cheap. Preferably the microphone is able to
automatically filter out, preferably electronically, undesired frequencies
(for
example according to the speed of the phase shifts, and/or the speed of any
other
detected distortions), so that for example very slow phase shifts such as
those
caused for example by wind or breathing or other air flows are preferably
ignored.
Preferably the transmitters and/or the sensors are as small as possible in
their own
physical size, in order to enable altogether a smaller microphone and/or for
more
exact analysis of the ultrasound peaks and/or forms.
CA 02446344 2003-10-23
23110103 Boris Dechovich & Yaron Mayer 17/39
Referring to Fig. 3, we show an illustration of a phase-shift view on a scope.
Since
the phase-shifts caused by the sounds in the described embodiments are
typically
very small and fast, it is difficult to see them on a scope. PIowever, by
using as high
as possible ultrasound frequency and thus increasing the number of ultrasound
wave
peaks within the preferably small gap between the transmitter and the sensor,
it is
possible to take the sum of the phase shifts and measure it very accurately
and very
fast.
Referring to Figs. 4a-b, we show two preferable examples of using, instead of
minute membranes, a group or matrix of minute elements that are preferably
more
rigid and can move more freely. Both figures show a sideways cross-section. In
Fig.
4a each preferably minute element is preferably shaped for example like two
round
or rectangular or for example 6-sided surfaces {41a & 41b) connected in the
middle
by a small rod or arm (44) and axe limited in their displacement range by a
blocking
narrower tunnel {43). The element can either be for example within a wider
tunnel
(42) or for example without it, in which case preferably the tunnel 43 is
preferably
longer in order to give it more stability. In Fig. 4b the small preferably
rigid element
is within tunnel 42 and is blocked at the two ends of its displacement path
for
example by mesh or wire structures (43) or for example narrower openings at
both
ends of tunnel 42, that allow free air flow but don't allow the vibrated
element to
escape. Another possible variation is that only one such block is needed,
since at the
other end is the base or the chip so the element cannot escape in that
direction
anyway. Since the vibrating elements are not entirely stable in the up-down
direction, this can make the sound less directional, and over a large array or
matrix
of such elements the random sideways fluctuations can make the sound emanate
over a wider angle. Of course these are just two examples and many other
variations
and configurations can also be used for enabling the more freely movable
elements.
Referring to Figs. 5 a-g we show side view cross sections of a few preferable
variations of using freely moveable, preferably fox example rectangular and/or
hexagonal and/or fractal shaped, smaller elements, preferably thin solid
plates (for
example thin aluminum foil or plastic), connected to one or more larger
elements,
which are vibrated by one or more for example electromagnetic coils or
capacitors
or Piezo elements or electronic taps or any other appropriate means, so that
the
entire bunch of elements vibrate together without having to apply a separate
transducer for each of them. This way all of the small elements can vibrate
simultaneously at a large number of frequencies, and there is no need to
explicitly
take care of cross-overs like among 3 membranes, since all the elements are
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 18/39
automatically activated as one group. The small elements can be for example
all of
the same size, or with various sizes, for example with the larger ones more in
the
center. The number of vibrated elements can be for example any convenient
number between 3 to a few dozens or a hundred or more. Fig. 5a shows fox
example
a hierarchical structure (51) which supports multiple small preferably thin
solid
plates (52). Of course, this is just a side-view cross section so from above
it can look
for example more like a checkerboard. Each branch in the hierarchical
structure can
have two or more sub branches, and can be connected to each sub-branch for
example in its middle or for example with some shift from the middle, in order
to
create also better vibrations at the periphery. The hierarchical structure
itself is
preferably based on small arms or needles and/or springs that go in all needed
directions. Another possible variation is that it is based on larger plates to
which
each time smaller plates are connected. The hierarchy depth can be for example
two
or more levels. Fig. 5b is similar to Fig, Sa, except that there is no
hierarchical
structure and the plates (52) are connected for example to some wires or mesh
structure {51) that preferably move through the air with little resistance.
Fig. 5c
shows a similar structure except that, instead of a hierarchy, multiple bent
or curved
needles or arms and/or springs (51 ) go sideways in all needed directions to
support
the multiple small elements (52). Preferably the needles or arms emanate from
one
center, but another possible variation is that for example more than one
center is
used. Figs. Sd-a are top views of other possible variations, where the
hierarchical
structure is based for example on a central larger plate in the center (52a)
and
smaller plates (52b) are attached to it preferably sideways (for example by
small
arms or points of connection or partial overlap), and yet smaller plates (52c)
are
preferably attached to each of the previous plates, etc., in a recursive
manner (For
simplicity of the drawing only some of the 52c type plates are shown in Fig.
5d).
The hierarchy can be for example of 2 or more levels, and each larger plate
can
expand into 1 or more additional directly attached plates, and the plates can,
again,
be for example round or rectangular, or hexagonal or fractal, or of any other
convenient shape. The plates can be all based on a similar shape, or for
example a
combination of different shapes. The plates can all be for example on one
common
plane or on more than one plane. In each step of the recursion the plates can
become
for example half the size of the previous plates, or any other convenient
ratio, or for
example become smaller by different ratios between levels of the recursion
and/or
within the same level. If for example rectangles or hexagons are used instead
of
circular plates, the arms or points that connect each plate to the smaller
plates that
branch from it can be for example from the middle of some edge, or from some
of
the corners, or for example with some shift from that. Another possible
variation is
CA 02446344 2003-10-23
23/I0/03 Boris Dechovich & Yaron Mayer 19/39
that at least some of these separate elements are connected, preferably with
one or
more point-connections, for example by strings, to a periphery and/or also to
each
other in addition to or preferably instead of solid connecting arms among
them. This
way they all become for example like nodes on a net, so that they can have at
the
same time the behavior of a large membrane and also have much more freedom for
each part of it, which can behave like a small membrane. The idea of using for
example a mixture of solid arms and strings, so that for example in some of
the
connections the solid arms are used and in others strings are used, has the
further
advantage that the solid arms can transfer for example more the vibrations of
higher
frequencies and the strings can transfer also the vibrations of lower
frequencies.
Another possible variation is to use also, in addition or instead, for example
also
some springs or combination with springs. Another possible variation is to use
in
this case for example membranes instead of the solid elements. Another
possible
variation is to connect for example only minute elerrlents or minute membranes
in
this net-like way, so that for example all elements are Iess than 1 cm is
size, but
together they can all behave also Iike various larger elements would react
while
maintaining also their much smaller vibrations. In all of Figs. Sa-a
preferably the
connection to each small element is either at its center on with some shifting
from it
or for example at the side. Preferably in all of these variations the gaps
among the
plates are as small as possible but without being so small that they can hit
each other
while vibrating. Since all the arms have relatively free movement from each
other,
the small elements can vibrate in a number of directions, thus creating sounds
in
multiple directions, as is important for loudspeakers. 'fhe plates themselves
can be
for example in one common plane or in a number of planes, thus allowing even
some overlap in the surface area they cover or allowing them to be with less
gaps
among them on the surface and still with less chance of them hitting each
other
while vibrating. Another possible variation, as explained in clause 4 of the
patent
summary, is that the plates are not all facing the same directions, so that
for example
they are in a convex formation (and/or concave) and/or in a wavy orientation
and/or
for example a hyperbolic reflector is used, in order to broadcast sounds
better in
multiple directions. The multiple orientations can be accomplished for example
by
adding some bends in the supporting arms and/or in at least some of the
plates.
Another possible variation is to connect for example at least some of the
elements
for example in just 2 poinis to strings - which means that they can also
rotate freely
around the axis of connection - so that multiple directions can also be
randomly
generated. Another possible variation is to use for e~;ample various lengths
of the
strings for various elements, so that the various lengths also contribute to
additional
combinations of frequencies. Another possible variation in any of these
CA 02446344 2003-10-23
23/10!03 Boris Dechovich & Yaron Mayer 20/39
configurations is for example to add additional transducers inside the
hierarchy,
which additionally vibrate the elements and/or for example to use additional
transducers at various places on the base to which al/ the elements are
connected.
Another possible variation in any of the above variations is that the elements
are not
free on a single arm, but are for example each connected to some frame around
it for
example by a few preferably small points of contact (for example a point of
connection at the middle of each of its sides), as is the case in the above
interconnected elements variations, so as to enable more easily additional sub-
resonances within it, yet on the other hand still leaving it more free than
elements
that are connected on their circumference. Each point of contact can be for
example
based on a string or another preferably small needle or arm or a spring.
Another
possible variation is that each plate or element or at least one or some of
them have
a more complex shape at the edges, so as to enable more free vibrations in
various
frequencies, such as for example a Fractals-like shape in the corners and/or
the
edges, so that each corner for example branches into a few more smaller
corners
which then can also for example branch into a few more smaller corners or
edges,
etc. Fig. 5f shows one possible example of this, for example in the shape of a
typical
snowflake structure. ~f course, the recursion can continue further, so as to
get even
a more fine structure at the edges, as shown for example in Fig. 5g. Since at
20KHz
for example the wavelength is 1.7 cm, fine structures of for example !cm or a
few
mm or less at the smallest branches should work quite one. The additional
branches
can be for example all in one plane, or in more than one plane, so that some
branches are for example lower or higher than others. ~f course this is just
an
example and other recursive or fractal-like shapes can also be used. The
recursive
fractal-like elements can be used for example in combination with any of the
above
variations, and especially in the interconnected net variations, so that in
this case the
recursion is preferably both with each element and among the elements, and
thus not
only the entire structure but even each single element in it can efficiently
vibrate in a
large range of frequencies. Another possible variation is for example to use
some
combination of solid panels with soft panels or membranes (for example by
using a
softer element in the middle), so that for example the displacements or
flexible
movements of the soft panels and/or of the strings help also create multiple
directions since it changes the angles of the other elements. The net itself
can be for
example in a relatively flat speaker, or in a somewhat deeper speaker that
allows it
more wide displacements for example in the center. However, since for example
displacements of 2-3 cm or even less should be enough, preferably the speaker
is
still relatively flat, thus saving space and making it more convenient to use.
But its
fraetallic nature can make a plate constructed like this become a very good
speaker
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 21/39
even if used alone, preferably when connected only with one or a few point-
connections, and then of course the speaker can be even more flat. Another
possible
variation is that preferably the fractallic plate is actuated for example by
one or
more of the new Helimorph piezoceramic actuators by 1 Ltd (which is based on
two
or more layers of the piezoelectric material, which are surrounded and
separated by
conductive electrodes, and are then twisted into a spiral, and the spiral is
then
twisted into half a circle, thus creating very efficient linear movements), or
similar
type of actuator. This has the advantage that if the voltage remains constant
the
Helimorph device remains in a fixed position, and thus can automatically also
stop
the movement like a dumper. If the Helimorph or similar shape is used in the
embodiments where more than one such plates are connected recursively, then
for
example one such Helimorph can be used to actuate the whole set of plates, or
for
example the connections between the plates also use Helimorphs as the link,
and
preferably the Helimoprh connections are smaller for the smaller plates, and
thus for
example the smaller plates can preferably also stop vibrating automatically
when the
sound stops. Another possible variation is to use any of the above recursive
or
fractal-like structures or variations as another way of creating a wide-
frequency-
range microphone, by using a sensor instead of the transducer, since such
structures
can also respond better to various frequencies than for example a single
membrane
or a single plate. Another possible variation is to vibrate for example just 1
plate like
in the NXT technology, thus relying mainly on the various resonances and
standing
waves within that 1 plate, but preferably hold it freely on a small needle or
arm
instead of attaching it to a frame, or connect it with preferably only a few
point
connections on the periphery (for example just 3 or 4 point connections, for
example
with strings or small arms or springs, preferably at some points in the edges
and not
in the corners which have to vibrate more freely), and/or use elements with a
more
complicated structure than just four corners, such as for example with a
Fractal-like
complexity at the edges, and/or use more transducers on various places to
vibrate
the plate, as described above. However, connecting more than one such element
for
example in any of the combinations described above can work even better.
Another
possible variation is to add also a dumper for blocking an element or for
example a
network of elements from continuing the vibration too Iong for example after
the
sound, even when there is supposed to be silence, and especially if they enter
their
natural resonance frequency. This can be done for example by using a resonance
box, which by the internal air's resistance automatically acts also as a
dumper.
However its effect will be limited since it is much less airtight in this case
than with
a normal speaker. Another possible variation is for example using a preferably
strong force in one or more transducers for helping the elements come to a
stop as
CA 02446344 2003-10-23
23/10/03 Boris Dechovich & Yaron Mayer 22/39
fast as possible when needed. Also, preferably the elements and/or structures
are so
designed so that their natural resonance frequency is not a problem. Another
possible variation is to use for example a one or more individual such fractal-
like
elements, for example of various sizes, without connecting them together, but
again,
preferably with just 3 or more point connections to their supporting frame, so
that
they work in synchrony preferably by a common control logic of their
transducers.
~f course, in this case also each of the plates can be for example actuated
individually by its own Helimorph (or similar device, and preferably the
Helimorphs size is correspondingly smaller for smaller plates and larger for
larger
plates. The transducers themselves can be driven for example analogically or
digitally. Various combinations of the above and other variations can also be
used.
~f course these are a just few examples and many other variations and
configurations can also be used for enabling the more freely movable elements.
While the invention has been described with respect to a limited number of
embodiments, it will be appreciated that many variations, modifications,
expansions and other applications of the invention may be made which are
included within the scope of tire present invention, as would be obvious to
those
skilled in the art.
